whereismyai
Agent Middleware
Agent Middleware is an API gateway that secures agent tool calls with signed permits, idempotency keys, and receipts.
Agent Middleware is a gateway that sits in front of an agent's risky tools such as refunds, payouts, and deployments. It ensures each action requires a signed permit, runs at most once per request key, and produces a signed receipt that can be checked offline. The system uses concepts like permits, calls, request keys, receipts, and internal credits for metering. It provides Python and TypeScript clients, framework wrappers, adapters, and a local gateway implementation. The gateway is designed for developers building AI agents that execute consequential actions.
Agent Middleware at a glance
Key features
- Gateway for risky agent tools
- Signed permits for action limits
- Request key idempotency handling
- Signed receipts for call outcomes
- Python and TypeScript SDKs
- Local gateway via Makefile
Use it for
- Securing agent tool calls for refunds and payouts
- Verifying agent action receipts offline
- Managing agent budgets and expiration limits
Alternatives to Agent Middleware
- Agent2api — It combines multiple AI desktop client accounts into one local OpenAI-compatible API.
- Adyen — Share AI agent resources across coding tools and repositories easily.
- agentgateway — It unifies service, LLM, and MCP traffic into one open-source binary.
- Composio — Equips AI agents with tools, managed user auth, and sandboxes across 1,500+ apps
- AgentGate — Blocks attacks in under 35ms with a local neural classification head.
- APIStack — Connect AI agents to over 100 tools using one API key or a single MCP endpoint.
- MyAgent — It features built-in permission controls, long-term memory, and skill self-evolution.
- Melra — Governs AI agent tool calls with local policies, approval gates, and verified receipts.
- Hermes WebUI — Run one server and access your agent from any device.
- Proxy — Restricts agent access and stops context sharing across multiple personal agents
- AgentTrust — Enforces zero-trust policy rules and maintains an immutable cryptographic audit ledger.
- BrowserOS — Runs entirely locally and lets AI agents use real user logins to complete tasks.
All Agent Middleware alternatives